Most of you explained the advantages of studying Cepaea but many did not explain the bigger picture: what were we interested in. After all, we are not (most of us) actually interested in snails!! The experiment was designed to tease apart the effects of genetic drift, gene flow and selection in explaining differences in allele frequency between populations - to learn lessons that can be applied to all species including humans. An really good intro would have briefly explained why we might be interested in doing that too!
Some of you were confused about the effects of gene flow: it reduces differences in allele frequency between adjacent populations. It may also increase the genetic diversity of populations by introducing alleles from neighboring populations... especially if there is selection for different alleles in those populations.
Many of you included material that would be expected in the methods section. You needed to explain the overall logic of your experimental design, not the details of what you did. Similarly, the details of statistical analysis belong in the results or methods, and interpretation of your actual results belongs in the discussion.
One attribute of the snail is small dispersal distance. This means that genetic patterns will be localized, and hence occur on a scale amenable to study. Just because the distance travelled in one generation might be 20m, that does not mean populations separated by 20m will be unaffected by gene flow. Genes will move over successive generations between those populations.
A difference in morph frequency between populations does not indicated that gene flow between them has reduced, only that is it must be insufficiently large to have homogenized allele frequencies. The gene flow could have increased, decreased or stayed the same!
You needed to explain that your experimental design controlled for some factors (e.g. for altitude, by having samples at the same height) and investigated others (e.g. the difference between bushes and open ground). (or you could have done it the other way round, all in grassland but different heights)
In many of your reports the basic logic for distinguishing the effects of selection and drift needed spelling out. In particular, you should have pointed out that both of them could lead to differences between samples taken from grassland and (say) woodland. Drift would be implicated if the woodland samples were just as different from each other as woodland to grassland. Selection if the wooldland ones were similar, the grassland were similar but woodland was different from grassland.
A more sophisticated answer would have pointed out the problem that (for example) grassland sites might actually differ in ecology, in ways which we have not detected. Hence we might inadvertently attribute the results of selection to drift.
If you define drift and selection get it right!
Polymorphism is a property of a population, not an individual
Drift accumulates over many years. We are not looking at differences one year to the next.
We do observe fixation. All the populations were polymorphic (weren't they?).
Do not use the word 'this' without qualification.... it makes the reader work too hard.
E.g. The reason for this may be drift (unclear)
The reason for this higher frequency of yellow may be drift (clearer)
Selection drift and gene flow do not explain the phenotype… they explain different phenotype frequencies in different populations.
Predation is not different from selection, it is one form of selection.
If your design had 4 populations taken from four different habitats differences could be due to drift or selection - it will not be possible to tease the effects of the two processes apart.
The simple genotype-phenotype relationship found in the snail does not mean drift is easy to detect, only that differences in allele frequency can be detected.
Doing your analysis (do this only after we have done the lecture on Chi square analysis)
You have to make your own decisions about how to draw up the tables, because the number of tests you do, the categories of snail you compare and the populations you compare will all depend on what you think are biologically relevant questions. There is no single correct answer.
Do you think the number of yellow vs the number of pink is important?
In which case you might draw up a table like this to see if there is any significant difference between any of your populations.
Yellow
Pink
Pop1
Pop2
Pop3
Pop4
and do a contingency table Chi squared test as set out on the sheet.
See also here
Alternatively you might think that there are 3 important categories of snail:
banded or Brown
pink,
unbanded yellow
and want to see if they differ in the uphill vs downhill populations.
Lets assume that Pop1 (uphill) and Pop 2 (downhill) were one pair of populations close to each other
and Pop3 (uphill) and Pop4 (downhill) were another pair close to each other but distant from the first
You might then conduct two tests
5 Banded or brown
Pink .
Unbanded yellow
Other .
Pop1
Pop2
and
5 Banded or brown
Pink .
Unbanded yellow
Other .
Pop3
Pop4
So you can see - it all depends on the details of your experiment and what you think the biologically important comparisons are.
Most of you explained the advantages of studying Cepaea but many did not explain the bigger picture: what were we interested in. After all, we are not (most of us) actually interested in snails!! The experiment was designed to tease apart the effects of genetic drift, gene flow and selection in explaining differences in allele frequency between populations - to learn lessons that can be applied to all species including humans. An really good intro would have briefly explained why we might be interested in doing that too!
Some of you were confused about the effects of gene flow: it reduces differences in allele frequency between adjacent populations. It may also increase the genetic diversity of populations by introducing alleles from neighboring populations... especially if there is selection for different alleles in those populations.
Many of you included material that would be expected in the methods section. You needed to explain the overall logic of your experimental design, not the details of what you did. Similarly, the details of statistical analysis belong in the results or methods, and interpretation of your actual results belongs in the discussion.
One attribute of the snail is small dispersal distance. This means that genetic patterns will be localized, and hence occur on a scale amenable to study. Just because the distance travelled in one generation might be 20m, that does not mean populations separated by 20m will be unaffected by gene flow. Genes will move over successive generations between those populations.
A difference in morph frequency between populations does not indicated that gene flow between them has reduced, only that is it must be insufficiently large to have homogenized allele frequencies. The gene flow could have increased, decreased or stayed the same!
You needed to explain that your experimental design controlled for some factors (e.g. for altitude, by having samples at the same height) and investigated others (e.g. the difference between bushes and open ground). (or you could have done it the other way round, all in grassland but different heights)
In many of your reports the basic logic for distinguishing the effects of selection and drift needed spelling out. In particular, you should have pointed out that both of them could lead to differences between samples taken from grassland and (say) woodland. Drift would be implicated if the woodland samples were just as different from each other as woodland to grassland. Selection if the wooldland ones were similar, the grassland were similar but woodland was different from grassland.
A more sophisticated answer would have pointed out the problem that (for example) grassland sites might actually differ in ecology, in ways which we have not detected. Hence we might inadvertently attribute the results of selection to drift.
If you define drift and selection get it right!
Polymorphism is a property of a population, not an individual
Drift accumulates over many years. We are not looking at differences one year to the next.
We do observe fixation. All the populations were polymorphic (weren't they?).
Do not use the word 'this' without qualification.... it makes the reader work too hard.
E.g. The reason for this may be drift (unclear)
The reason for this higher frequency of yellow may be drift (clearer)
Selection drift and gene flow do not explain the phenotype… they explain different phenotype frequencies in different populations.
Predation is not different from selection, it is one form of selection.
If your design had 4 populations taken from four different habitats differences could be due to drift or selection - it will not be possible to tease the effects of the two processes apart.
The simple genotype-phenotype relationship found in the snail does not mean drift is easy to detect, only that differences in allele frequency can be detected.
Doing your analysis (do this only after we have done the lecture on Chi square analysis)
You have to make your own decisions about how to draw up the tables, because the number of tests you do, the categories of snail you compare and the populations you compare will all depend on what you think are biologically relevant questions. There is no single correct answer.
Do you think the number of yellow vs the number of pink is important?
In which case you might draw up a table like this to see if there is any significant difference between any of your populations.
and do a contingency table Chi squared test as set out on the sheet.
See also here
Alternatively you might think that there are 3 important categories of snail:
banded or Brown
pink,
unbanded yellow
and want to see if they differ in the uphill vs downhill populations.
Lets assume that Pop1 (uphill) and Pop 2 (downhill) were one pair of populations close to each other
and Pop3 (uphill) and Pop4 (downhill) were another pair close to each other but distant from the first
You might then conduct two tests
and
So you can see - it all depends on the details of your experiment and what you think the biologically important comparisons are.